RAID: What It Is and What It Means for Your Organization
As any IT professional can tell you, the tech world is full of acronyms. Some of them are common knowledge to the layperson, but others are much less familiar.
In a previous blog post, we looked at the question of whether or not RAID is necessary for solid state drives. In this post, we’ll take a deeper dive into what RAID is and why it matters for your organization. Keep reading to learn more.
What is RAID?
RAID stands for Redundant Array of Independent Disks. But what exactly does this mean?
Essentially, RAID is a means of accounting for the inevitability of hard drive failure. Like any other piece of hardware, no hard drive is going to last forever. (Incidentally, this is one reason why purchasing a refurbished hard drive can be a cost effective and practical approach to augmenting your current storage: a refurbished drive can last nearly as long as a brand new one.) When a drive does finally fail, what happens to all of the data stored on it? And what about the operating system that the drive is running?
RAID was introduced as a file storage system in order to avoid a total catastrophe in the event of a single drive failure. Rather than storing data in a single place, RAID involves storing the same data in different locations across multiple hard drives. This acts as a safeguard. When a single drive fails, there are backup copies of that same data available.
RAID Levels: Balancing Security and Performance
The reality of RAID is a bit more complicated than the simple description above. There are actually different levels of RAID available, each with its own advantages and disadvantages.
RAID 0 is the highest performance but least secure form of RAID. This is due to the fact that it involves disk striping -- the act of splitting data for individual files across a number of disks, thus allowing increased data throughput -- without any actual data redundancy.
The next level, RAID 1, involved what’s called disk mirroring. This means that data is written to and read from identical drive pairs. This is one of the most straightforward approaches to data redundancy with RAID, as there’s a backup disk available if one of the disks in a pair ever fails.
RAID 5 uses striping with parity. This involves striping data across multiple disks (as with RAID 0) along with what’s called parity information, or tiny amounts of data that can be used to reference bigger chunks of data. In the event of a drive failure, parity information can be used to recover data.
RAID 6 is similar to RAID 5, except it offers striping with an additional parity block. This offers increased redundancy at the cost of additional storage space.
Finally RAID 10 consists of both striping and mirroring. It’s essentially a combination of both RAID 1 and RAID 0. While this might sound ideal, the tradeoff is that the amount of physical storage required can be cost prohibitive.
Hardware vs Software RAID
When it comes to setting up RAID, there are two approaches you can take: software and hardware RAID.
With software RAID, the RAID disks attached to your servers are turned into RAID arrays using the processing power of the server’s motherboard. This is carried out via whatever operating system the server is running.
In the case of hardware RAID, a separate controller card (either inside the server or within the storage system) is used to conduct the actual tasks associated with writing data. This removes the extra load from your server, but at the added cost of a discrete controller card.
Hot Spares and RAID
As already mentioned above, the costs associated with cloud computing have declined in recent years. Regardless, though, there is still a cost associated with IaaS providers. These sorts of month-to-month pay plans are here to stay. And once you’ve migrated your data from your local network to the cloud, you’re essentially beholden to whatever fee structure your IaaS provider chooses to implement. Be sure to run the numbers ahead of time, as moving to the cloud won’t always result in cost savings.
Why Use RAID?
No matter how well maintained or new your hardware might be, all drives are prone to failure. RAID offers a means of backing up your data and ensuring that your system stays up and running, even if a drive fails.
Here at IGS, we offer a wide selection of used and refurbished PERC RAID controllers from Dell, as well as IBM ServeRAID controllers. All of our hardware is tested and inspected by our expert team of technicians, and we only sell refurbished items that meet our high quality standards.
Need help finding the right hardware for your organization? Chat with one of our associates now, or contact us for assistance.